Topics in Geriatric Rehabilitation • Volume 27, Number 2, 151–161 • Copyright © 2011 Wolters Kluwer Health | Lippincott Williams &

Wilkins
DOI: 10.1097/TGR.0b013e31821bfe68

Yoga-Based Maneuver Effectively Treats Rotator

Cuff Syndrome
Loren M. Fishman, MD; Allen N. Wilkins, MD; Tova Ovadia, PT; Caroline Konnoth, PT;
Bernard Rosner, PhD; Sarah Schmidhofer, BA, RYT

chrane studies find little evidence that either conservative

Objective: To measure efficacy of a simple maneuver in the
or surgical remediation is entirely successful1,2 though each
conservative treatment of rotator cuff syndrome. Design:
method reports gains.3-13 Because RCS is most common in
Before-and-after study with mean 30-month follow-up
the elderly people, and massive tears are frequently inop-
(range: 9 months–8 years). Setting: Private practice. Partici-
erable, an effective nonsurgical method is welcome. This
pants: Fifty consecutive outpatients with magnetic resonance
study focused on patients with supraspinatus tears, though
imaging–confirmed partial or full-thickness supraspinatus
other injuries were sometimes present.
tears. Intervention: A single partial weight-bearing maneuver
There is surprisingly little correlation between postsur-
involving triangular forearm support (TFS) was repeated in
gical tendon integrity and clinical improvement.6,7,10,13-23
physical therapy for a mean 5 sessions (range: 1 session–24
Physical therapy (PT) generally focuses on scapular stabil-
sessions). Main Outcome Measures: Maximal painless active
ity, kinesiological, and modality-oriented means of healing.
abduction and flexion before and after performing TFS, pain
In this article, we condense these conservative strategies
on maximal abduction and flexion before and after perform-
into a single exercise that was discovered serendipitously
ing TFS, and at mean 2.5-year follow-up. Results: Mean
and is effective in 30 seconds. After magnetic resonance
painless active abduction increased from 73.7 to 162.8
imaging (MRI) confirmation of a massive tear, one author
(P  .001; SD  32.3); mean painless active flexion
briefly practiced headstand during the month-long wait for
increased from 84.1 to 165.4 (P  .001; SD  36.7). In
surgical consultation (see Figure 1). Upon righting himself,
2.5 years follow-up mean combined painless abduction and
he experienced painless full abduction and flexion in the
flexion active range of motion was 171.5 (P  .001; SD 
arm that previously had less than 60 of either motion.
14.4). In immediate post-TFS testing and after 2.5 years
Subsequent electrodiagnostic examinations found the
mean visual analogue scale pain rating during maximal
subscapularis, rhomboides, serratus anterior, and pectora-
abduction and flexion fell from 5.46 to 0.97 (P  .001; SD 
lis muscles significantly more active during headstand and
2.6). Conclusions: These values compare favorably with most
in abduction and flexion immediately thereafter. A safer
surgical and nonsurgical studies. Triangular forearm support
and simpler version of headstand, triangular forearm sup-
plus physical therapy appear to improve abduction and flex-
port (TFS), named the “Tova maneuver” after its physical
ion and reduce pain immediately and in the longer term after
therapist inventor, did the same thing (see Figures 2 and 3).
rotator cuff syndrome. Key words: conservative treatment,
The maneuver is adapted from B. K. S. Iyengar’s24 meth-
rotator cuff syndrome, triangular forearm support
od for headstand25 but is not universal in Yoga. Subscapu-
laris activity seems crucial for the benefits reported here.

S
urgery is often recommended for rotator cuff syn-
drome (RCS), but enthusiasm drops off abruptly
with massive tears, and in the elderly people. Co-
Author Affiliations: Columbia College of Physicians and Surgeons, New
York, New York (Dr Fishman); Harvard Medical School, Boston, Massa-
chusetts (Drs Wilkins and Rosner); Manhattan Physical Medicine and Re-
habilitation, New York, New York (Mss Ovadia and Konnoth); and Brown
Medical School, Providence, Rhode Island (Ms Schmidhofer).
The authors thank Dr David Palmieri, Dr Allan Cummings, Carol Stratten,
BSRT(R)(MR), and Norman Brettler of MHA Tilton Dynamic Imaging in
Northfield, New Jersey, Drs Jerald Zimmer and Alain D. Hyman of New York
Medical Imaging, Aveenash Chatterpaul of Doshi Diagnostics, both of New
York, for their assistance in obtaining and interpreting the MRI, CT, and ra-
diographic findings reported in this article, and the Clevemed company for
the Biocapture device used in Figures 9A and 9B. The authors also thank
Mikiko Murakami, medical student, and Michele Blacksberg, RN, the latter
of whom was compensated for assistance in collecting the data.
Correspondence: Loren M. Fishman, MD, Columbia College of Physicians
and Surgeons, 1009 Park Avenue, New York, NY 10028 (Loren@sciatica.org).
The steps in Tables 1 and 2 were not put together by Mr
Iyengar; he is not responsible for their use. See his book
for headstand itself.24
Inversion-positioning and muscular activation are well-
established aspects of standard PT, and there was no human
experimentation. Sound Shore Medical Center’s institutional
review board approved the larger study of which this is a part.
MATERIALS AND METHODS
Patient selection
Inclusion criteria:
1. Sudden reduction in painless range of abduction or
flexion.
2. MRI-confirmed tear of the supraspinatus, with or
without tear of the infraspinatus or teres minor.
Exclusion criteria:
1. Tear of the subscapularis.

Topics in Geriatric Rehabilitation www.topicsingeriatricrehabilitation.com 151

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 151 30/04/11 10:58 AM

 Figure 2. Cycle of chair-assisted headstand (Urdhva Dan-
dasana), following the directions in the text. The empha-
sis is on safety. The torso must be fairly close to vertical,
which translates into its being close to the chair. Patients
with cervical pathology, orthostatic hypotension, glauco-
ma, berry aneurisms, and other conditions contraindicat-
ing inversion should use the Tova maneuver. See Figure 3.

Figure 1. Magnetic resonance imaging confirming a com-

plete through-and-through tear of the supraspinatus, with
retraction of the tendon. Distal and proximal fragments in-
dicated by arrows (A). Same patient unhealed 3 years after
using triangular forearm support, but with continued full Figure 3. Hands are clasped behind head. The same con-
active abduction and flexion without pain (B). Progressive certed action of triangular forearm support may be recruit-
elevation of humeral head seen after 9 years, still with pain- ed through resisting the horizontal vector generated by
less full active abduction and flexion (C). this slanted position.

152 www.topicsingeriatricrehabilitation.com April–June 2011

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 152 30/04/11 10:58 AM

 TABLE 1 Verbal Directions for Inverted Triangular Forearm Support, Chair-Assisted
1. Stand with your back to a chair.
2. Place your right shin on the seat of the chair.
3. Bend forward, placing your palms on the (blanketed or carpeted) floor, fairly close to the chair.
4. Raise the left shin to the chair seat, so that you are now kneeling on the chair with your hands on the floor.
5. Bend your elbows as you lower your head to the floor. After your head is securely planted on the floor, center it between your
hands. Be careful not to place your weight either on the forehead or the back of your head, but rather at the fontanelles, the
spot that is soft in babies.
6. Make an equilateral triangle on the floor with the little finger side of your forearms by clasping your hands. Place the heels of
your hands, not the palms, in contact with the back of your head.
7. At this point, with your weight chiefly on your head, press down with your forearms to lift your shoulders away from your ears.
Widen and raise your shoulders further from the floor.
8. Stay in this position for 30 seconds.
9. Now bring first one knee, then the other to the floor, using your hands and arms for support as necessary.
10. After 5 to 10 seconds, quit the position and stand up normally.
11. Boldly raise your arms up to vertical. Do not stop at 90 and wait for it to hurt. Continue the motion as far as possible.
12. Do the same with flexion.

2. Paraesthesias, numbness, or pain radiating below Triangular forearm support

the mid humerus.
3. Neuromuscular disease such as stroke, multiple
sclerosis, or myopathy.
4. Previous shoulder injury or shoulder surgery.
5. Cervical pathology or other conditions contraindi-
cating axial pressure.
Clinical management
Full medical histories and physical examinations included
goniometric determination of painless active ranges of ab-
duction and flexion.26-28 Participants self-rated pain on the
visual analogue scale (VAS) during maximal abduction and
flexion.28 Participants were introduced to the idea of exer-
cise as treatment, and performed TFS during their first visit
(see Figures 2 and 3).
The authors and physical therapists demonstrated one
or both forms of TFS to participants as necessary (see
Figures 2, 3 and Tables 1, 2). Participants remained in
the inverted or slanted position for 30 seconds, a period
that was found both harmless and effective. After 30 sec-
onds, the inverted patients returned one leg and then the
other to the floor, and kneeling, raised their heads while
unclasping their hands. After 5 to 10 seconds on all fours
(long enough to avoid light-headedness, brief enough to
keep the current exercise in mind), they were helped to
their feet. Tova-maneuver patients had only a few seconds’
delay between their exercise and their post-TFS abduction
and flexion. Painless ranges of motion and pain at maxi-
mal ranges were then rerated. One post-TFS instruction
was given to all patients for both abduction and flexion:

TABLE 2 Verbal Directions for Diagonal Triangular Forearm Support Against a Wall
1. Interlock your fingers, making an equilateral triangle with your forearms as you place them against a wall.
2. Place the fontanelles in the center of that triangle.
3. Walk away from the wall, so that your torso now slants toward the wall. Some weight is now on your head.
4. Lower your chest and press your elbows and forearms into the wall, using the pressure to pull your shoulders far away from the wall.
5. Draw your shoulder blades back, down and apart, still pressing against the wall with your elbows and forearms. Press your
shoulders, but not your head, away from the wall.
6. Stay like this for 30 seconds.
7. Now come away from the wall and stand up straight.
8. Boldly lift your arms up to vertical. Do not stop at 90 and wait for it to hurt.
9. Do the same with flexion.

Topics in Geriatric Rehabilitation www.topicsingeriatricrehabilitation.com 153

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 153 30/04/11 10:58 AM

 TABLE 3 Demographics of Subjects
Subjects Right-Sided Dominant Full-Thick- Other Months Previous
Demographics (n) Agea Women Tear Sided Tear ness Tears Pathologyb After Onsetc Yoga
All patients 49 62.9 33 26 31 37 25 24.5 18
SD 14.29 50.56
Range 32-97 1-268
a
When first seen.
b
For example, Infraspinatus tear, other tendinoses, labral tear.
c
Months after onset when first seen.

“Boldly lift your arms up to vertical. Do not stop at 90 and TABLE 4 Results
wait for it to hurt.” Twenty-three patients were taught the
Rotator Cuff Study—Efficacy Measures
chair-assisted headstand (Urdhva Dandâsana) version;
27 were taught the Tova maneuver. Active
Total ROM Mean  SD t Pa
Follow-up Mean ROM 78.5  25.8
Patients were prescribed 2 to 3 weekly PT sessions for 6 0.30 .77
When first seen (n  46)
weeks, receiving standard PT for RCS as outlined on page
171.5  14.4
6, and daily TFS practice. Painless range of motion and Best mean ROM in follow-up
(n  48)
1.91 .09
pain at maximal ranges were examined weekly in PT and
94.0  27.4
at medical visits at 6 weeks, 3 months, 1 year, and annually Difference in ROM 2.47 .02
(n  46)
thereafter. Phone calls were used when necessary.
P valueb (before vs after) .001
Statistical analysis Abduction
The statistical consultant used the Wilcoxon rank sum and
signed-rank tests to compare initial ranges of abduction and 73.6  24.4
Abduction before 0.30 .77
(n  48)
flexion and VAS pre- and post-TFS and at final follow-up. These
nonparametric tests were used to assess results in the advent 162.8  24.7
Abduction after 1.58 .12
of a nonnormal distribution of the data. Paired t tests and (n  48)
2-sample t tests analyzed pre-TFS versus final ranges of motion. 89.3  32.3
Abduction difference 0.35 .73
(n  47)
Source of funding
There was no external source of funding for this study. P valueb (before vs after) .001
Flexion
RESULTS 84.1  33.3
Patient data Flexion before 0.06 .95
(n  43)
Fifty patients qualified for the study. One patient elected to 165.4  19.4
have surgery, leaving 49 patients. The group included 16 Flexion after 0.80 .43
(n  44)
men (32.7%), 37 full-thickness tears (FTT) (75.5%), involv-
81.7  36.7
ing 26 (53.1%) dominant extremities. Mean initial age was Flexion difference 0.28 .79
(n  42)
62.9 years (range: 32–97) (see Table 3).
P valueb (before vs after) .001
There were 11 additional tears to other muscles of the
rotator cuff; 7 patients had tendinosis, 3 had teres minor Pain with maximal abduction, flexion
atrophy, 3 had labral tears, 2 had bursitis. For 4 patients, 5.5  2.4
MRI suggested impingement.29 Mean prior symptom dura- Pain before (n  43) .005c
tion was 24.9 months. These statistics are similar to other 4.3  2.6
studies worldwide.3-13,15–23,29-42 Eighteen participants (36.7%) Pain difference (n  42) .36 c
had some previous experience with Yoga.
P value (before vs after) .001d
Mean painless and maximal abduction and flexion im-
Abbreviation: ROM, range of motion.
proved significantly immediately after the initial 30-second a
P value by 2-sample t test.
TFS (see Table 4). b
P value by paired t test.
1. Mean painless abduction, initially at 73.0 im- c
P value by Wilcoxon rank sum test.
proved to 162 immediately after TFS (P  .001; d
P value by Wilcoxon signed rank test.
SD  32.7) (see Figure 4).

154 www.topicsingeriatricrehabilitation.com April–June 2011

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 154 30/04/11 10:58 AM

 Figure 4. Active painless abduction and
flexion ranges of motion before and af-
ter 30 seconds triangular forearm sup-
port (TFS), and visual analogue scale at
maximal abduction and flexion before
and after 30 seconds TFS.

2. Mean painless flexion rose from initial 84.2 to
165.4 immediately after TFS (P .001; SD 
37.2) (see Figure 4).
3. Mean pain on maximal abduction and maxi-
mal flexion post-TFS (taking the higher score)
dropped from 5.46 to 0.97 or 4.49 points on the
VAS (81%), immediately after TFS (P .001; SD 
2.6) (see Figure 4).
4. Painless range of motion improved 100% or more
in 37 of 49 patients (75.5%) directly following
TFS. Mean improvement was 150% (SD  1.3;
median improvement  120%) (see Figure 5).
5. Gains were sustained in mean 2.5 year follow-up
(see Figure 4).
6. Three patients did not improve.
DISCUSSION
In an immediately beneficial intervention such as this, the
patients themselves supply the baseline and, in this sense,
are their own controls. Other conservative and surgical

Figure 5. Percentage and distribution

of improvement in abduction and
flexion after 30 seconds triangular
forearm support.

Topics in Geriatric Rehabilitation www.topicsingeriatricrehabilitation.com 155

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 155 30/04/11 10:58 AM

 Degeneration investigations were reviewed to estimate the comparative

Increasedf Increasedf
efficacy of TFS over the longer term.
Joint

Increased Increased

Increased Increased
Conservative therapies

NA
One recent conservative approach to FTT improved mean
Change

abduction and flexion by 23.7% and 14.4%, respectively, af-

Fatty

NA
ter 6 months of therapy.14 Tears were less extensive than
those of the current study, but end-stage improvement
of 131.4 and 144.6 were less than our study’s 162.5 and
Pain Pain Glenohumeral

Decreased

Decreased

Increasedf
Distance

165.4, respectively41 (see Table 5 and Figure 6).

NA

Zingg et al30 studied nonoperative care of massive RCS

in 40 patients for 4 years. Mean age was 64 (54-79). Mean
flexion improved from 115 to 139 (20.9%); mean abduc-
Before After

.97e
2.3

2.3

NA

tion rose from 118 to 140 (18.6%). Converting from that

study’s 15-point VAS to the 10-point VAS, pain after 4 years
averaged 3.5. No pretreatment values were given. Final
5.46
NA

NA

NA

VAS score in our 49 patients was 0.97 Mean glenohumeral

Conservative Studies of Zingg and Baydar Measured Passive Ranges of Motiona

separation decreased 2.6 mm in the study of Zingg et al;

Unchanged
Tear Size

Increased

Increased

tear size and fatty infiltration increased significantly. The

NA

number of irreparable tears doubled from 4 to 8 in 4 years.

In 30 months’ mean follow-up, our patients, including
3 with massive tears, had no new tears. Seven of 10 follow-
Change Years

up MRIs 0.3 to 6 years after treatment revealed accelerated

0.5

2.5
f/u

degeneration of the humeral head, in our patients, but all

maintained their original increases in degrees of painless
20.9

14.4

96.4

abduction and flexion. Over the mean 30 months study pe-

%

riod, VAS remained stable at 0.97 (see Table 5 and Figure 6).
No Change
Flexion

Surgery
165.4e
165.4
After

139d

Open surgical and arthroscopic studies present inclusion

criteria, measurement scales and time frames that are rare-
Zingg et al30 studied massive tears, found no change in active ranges of motion over 4 years.

ly comparable. One Swiss study40 of 26 men and 24 women

Flexion
Before

144.6

84.2
115

(mean: 58.5 years) with mean 12-month prior symptoms

(range: 3-48 months) documented pain reduction that
converts to 7.27 and 3.27 on the 10-point VAS, or 55%
Change

17.8

reduction in pain; our study saw 81% pain reduction (see

123
23
%

Figure 7). This postsurgical study increased range of mo-

tion in 25% of patients at 6-year follow-up. The present
No Change
Abduction

study doubled mean painless active abduction and flexion

162.6e
After

139c

162

in 82% and 75.5% of patients, respectively, immediately and

after 2.5 years (see Figure 5). However, longer follow-up
brings improvement in some studies and deterioration in
Abduction
Before

One patient studied serially more than 7 years.

others, discouraging strict comparison.

131.7
118

73

Other recent surgical studies3,11 increased flexion 16%

in 22 FTT, and 51% in 20 partial thickness tears, averaging
Abbreviation: ROM, range of motion.

33.5% improvement to mean maximum in the high 150

Unknown

Six cases of flexion less than 90.

Patients

range in 2-year follow-up.

(n)

19

19

49

Cole et al7 studied arthroscopic repair of 49 rotator cuff

tears with 2-year follow-up. Patient age averaged 57 years
Baydar (passive

(active ROM)

(range: 34-80 years), mean presurgical abduction and flex-

Zingg (passive

Current study
Study When

Zingg (active
TABLE 5

ion was 121 and 136, respectively. The 6 months’ and 1

Measured

ROM)b

ROM)b

ROM)

P  .047
P  .001

year’s gains were below ours, but at 2-year follow-up, the

P  .07

mean abduction and flexion were 2.2 and 6.6 above our
results (see Table 6).
b

d
a

156 www.topicsingeriatricrehabilitation.com April–June 2011

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 156 30/04/11 10:58 AM

 TABLE 6 Improvement in 49 Rotator Cuff Tears After Arthroscopy
Ranges of Motion
Preoperative 6 mo P 1y
Flexion 136 159 0.15
Abduction 121 142 0.09
Cole’s rehabilitation was well-managed and extensive.
Postoperative slings accompanied Codman exercises for the
first 4 weeks. Passive range of motion to tolerance in flex-
ion with internal rotation limited to 40 was done with elbow
anterior to the midaxillary line, and active assisted range of
motion at 4 weeks. Deltoid and biceps strengthening began
after 6 weeks. Weeks 9 to 12 stressed scapular stabilization ex-
ercises and posterior capsule stretching. With the exception
of Cole’s patients abating sporting activities for 4 to 6 months,
this therapy was similar to our own. However, our patients
had no “down time” whatever, and most were able to pursue
all normal activities 30 seconds after treatment onset.
Two other 2-year arthroscopic studies improved flex-
ion and abduction from 135 to 14913 and from 142 to
174,10 respectively. The first study examined repair of FTT
in patients of average age 60.7 years. The second study’s
patients averaged 58.3 years of age, with mean tear size of
2.47 cm, and had similar but more abbreviated postopera-
tive rehabilitation (see Figure 8).
Many other published studies are in the low to middle
range of these surgical studies.8,10,13,15,21 Though high pa-
tient satisfaction is frequently reported,8,10,11,15,29,33 also re-
ported are rerupture of the repaired tendon, postoperative
weakness, bone graft death, and infection.7, 22, 33,35-38,42-44
Puzzling consistent anomaly
Retear rates after surgery were generally in the 12%
to 32% range but some papers describe them as “very
high.”7,10,12,22,33,39,40,42-44 Curiously, satisfaction rates and ac-
tual ranges of motion do not correlate with these postop-
erative events. Many studies find no increased elevation of
the humeral head, nor deficits in ranges of motion with
retear,7,10,17-20,39 except for large recurrent tears. One 32-
patient, mean 31-month follow-up of bone graft surgeries
Figure 6. Comparison of current study with
other conservative studies. Visual analogue
scale multiplied 10 for all studies to fit scale
of graph.
Topics in Geriatric Rehabilitation
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
TGR2702-09.indd 157
P 2y P
162 .014 172 .0001
153 .006 165 .001
for massive tears reported MRI-confirmed graft death in all
15 cases.5 Nevertheless, patient satisfaction was high, and
range of motion improved with mean 30 to 35!5
Again, arthroscopic and open surgery is significantly less
likely to succeed in patients older than 65 years, or with
boney defects.3,5,7,10,14,15,20,22,34, Yet, despite these histological
failures, most patients with FTT or partial thickness tears
seem to improve.33,17-23,42 One study42 with a high retear rate
after massive tears had mystifyingly positive patient satis-
faction. Many found little correlation between tendon in-
tegrity and patient satisfaction.20,22,23,33,44
Furthermore, this first study showed less improvement
and higher percentages of retears when a shorter time period
elapsed between tear onset and surgery.45 Given increased fat-
ty infiltration and tissue deterioration over time, one would ex-
pect just the reverse, unless more than physical structure is in-
volved.46 Something beside tissue health may be relevant here,
something deeper beneath the surface even than the surgery.
It seems that RCS patients, with or without surgery, of-
ten inadvertently self-train to use a different set of muscles
for abduction and flexion, sparing themselves the pain and
disability that arises with contracting the torn supraspinatus
muscle. Triangular forearm support may give that training
to patients almost unknowingly in a very short period of
time. Our electrodiagnostic studies seem to confirm this.
How does TFS work?
One TFS patient had previous RCS surgery on the contralat-
eral shoulder. We performed 8-channel electromyography
on both sides’ shoulder girdle muscles during abduction.
We also compared electromyography of the supraspina-
tus and subscapularis of 3 RCS patients’ abduction before,
during and after TFS. The results were viewed by 2 blinded
physicians who scored electrical activity from 0 to 4.
www.topicsingeriatricrehabilitation.com 157
30/04/11 10:58 AM
 head of the humerus into the glenoid fossa.47,49-51 The ac-
tion might be compared to pulling horizontally on a slack
clothesline. As the rope tightens, the clothesline will rise
nearly to 90. Further pressure will tighten the line, but it
will never rise beyond the horizontal.
Several researchers have found major contributions
of the subscapularis, infraspinatus, and teres minor dur-
ing arm elevation.48-51 Triangular forearm support, moving
the shoulders caudad and retracting them, activates these
muscles. How does that help abduction?
The subscapularis appears actually to lower the humeral
head. The scapula itself is held fast to the spine by powerful
contraction of the rhomboids and serratus anterior,45-48,50,51 as
the humerus approaches 80. At that point the subscapularis,
exerts downward force on the head of the humerus in the gle-
noid fossa, pulling it caudad, away from the acromion. Lower-
ing the head tilts the shaft upward enough, creating a more
acute angle with the acromion, enabling the deltoid to con-
tinue the abduction or flexion toward vertical. The teres minor
balances the subscapularis’s external rotation (see Figure 9).
Figure 7. Comparison between surgical study (Linthoudt), Several studies report up to 85% greater subscapularis
arthroscopic study (Cole), and current study is only ap- activity in abduction in RCS.45,47,50,51 Opposing gravity with
proximate due to different metrics and different scales. the inverted or slanting body’s weight presents the sub-
scapularis with a more challenging foil against which it may
contract even more vigorously (see Figures 9A, 9B).
On all occasions, the same muscles, the subscapular- Triangular forearm support appears to reverse the roles
is, and to a lesser extent, the anterior and lateral deltoid, of key muscles. Between 80 and 110 the deltoid stabilizes
rhomboid major, serratus anterior, and pectoralis were the humerus, rather than lifting it, while the subscapularis
additionally activated during TFS, and when the subject ab-
ducted the affected arm in an upright position immediately
following TFS (see Table 7 and Figure 9). TABLE 7 Numbers Represent
During normal shoulder girdle abduction and flexion, the Electrophysiological Activity in
deltoid raises the arm to approximately 80, at which point its These Muscles During Abduction
force is directed nearly horizontally.45-52 In typical function, the and Flexion Before and Immediately
deltoid relaxes somewhat as humeral abduction or flexion ap- After Triangular Forearm Supporta
proaches the 80 mark, when the supraspinatus begins raising
Muscle Activity of Rotator Cuff Syndrome Patients During
the arm the next 20 to 40. At that elevation, a sufficient angle
Abduction Before and Following TFS
above the horizontal is formed between the acromion and the
humerus, creating a vertical vector in the deltoid’s pull that is Abduction Abduction
sufficient to resume lifting the arm47-50 (see Figure 9). Muscle Before TFS After TFS
When the supraspinatus is torn, the deltoid activity Rhomboid major 3 4
near 90 of abduction or flexion painfully compresses the Deltoid 4 4
Subscapularis 1 4
Rhomboid minor 1 3
Teres minor 1 2
Pectoralis major/
1 2
minor
Serratus anterior 1 2
Latissimus dorsi 0 0
Abbreviation: TFS, triangular forearm support.
Figure 8. Active flexion after arthroscopy and triangular a
Numbers represent the amount of electrophysiological activity 10 seconds after
needle insertion, at 90 abduction.
forearm support.

158 www.topicsingeriatricrehabilitation.com April–June 2011

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 158 30/04/11 10:58 AM

 Figure 9. Composite of 10 trials: Note difference in subscapularis and deltoid activity from 70 to 110 in abduction after
surgery (A) versus triangular forearm support (TFS) (B). C, The rhomboids stabilize the scapula as the subscapularis pulls
the head of humerus away from the acromion in post-TFS abduction. The usual roles of some shoulder muscles reverse:
the deltoid holds the humerus as a fulcrum while the subscapularis lowers the head, pivoting the humerus around the
still point at the deltoid’s insertion which, like a see-saw, lifts the arm until a sufficient angle forms to enable the deltoid
to continue the abduction.

and other muscles elevate the humeral shaft by depressing
the humeral head in a kind of see-saw motion. Caudad ten-
sion on the head of the humerus, learned in TFS, uses the
deltoid as a dynamic fulcrum that briefly steadies the proxi-
mal humeral shaft, while the downward pull of the sub-
scapularis lowers the humeral head enough to cantilever
its shaft upward. Then, the deltoid resumes abduction and/
or flexion. The suscapularis and other muscles continue to
exert some caudad pressure on the humeral head, avoid-
ing contact with the glenoid or acromion.
One successfully remediated patient remained in TFS
for the duration of a horizontal field 0.6 T Fonar MRI. Sub-
sequent computed tomography multidetector (64 detec-
tor) isotropic voxels in double angled multiplanar format-
ting with low table pitch (0.6 mm) confirmed the location
and form of the subscapularis during the maneuver. It was
read as extremely active.
Limitations and suggestions
Triangular forearm support
The best scenario would be if TFS rendered the supraspi-
natus completely inactive during abduction and flexion,
enabling its tendon to heal. However, the tendons of the 10
patients having subsequent MRIs were not healed. Some
showed substantial progressive arthritis. One patient with
massive tear had MRI and computed tomography 6 years
after TFS showing a high-riding humeral head with signifi-
cant arthritis, but with visible lowering of the humeral head
with (painless full) abduction (see Figures 1C and 10).
The idea that nonhealing is due to continued supraspina-
tus activation is supported in the literature.45-51 If true, paralyz-
ing the supraspinatus muscle while the subscapularis mecha-
nism is at work may be clinically useful. We recently began
administering botulinum neurotoxin to the supraspinatus
after teaching TFS in the second phase of the institutional
review board–approved study. Because the supraspinatus is
not, apparently, useful in post-TFS abduction and flexion, nor
in the puzzling studies reviewed earlier, temporary paralysis
inhibits no function. As the paralytic effect of botulinum neu-
rotoxin wears off after 8 to 12 weeks, a second MRI will con-
firm or disappoint the hope of healing.
This study
Study weaknesses include a relatively short follow-up time,
small patient numbers, consecutive sample without random-
ized matched controls, few postintervention measures, and

Topics in Geriatric Rehabilitation www.topicsingeriatricrehabilitation.com 159

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

TGR2702-09.indd 159 30/04/11 10:58 AM


Figure 10. X-ray study 6 years following triangular forearm
support in a 68-year old with massive left rotator cuff syn-
drome and earlier surgical revision of dominant right su-
praspinatus tear. High riding left humeral head and gleno-
acromial pseudoarthrosis (A) nevertheless allows painless
full range of abduction through downward pull on humer-
al head with abduction. B, C, The deltoid raises the arm on
the left to 90, but the horizontal vector of the deltoid will
not raise the arm on the right beyond 90 without help.
160 www.topicsingeriatricrehabilitation.com
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
TGR2702-09.indd 160
few postintervention MRIs. These flaws obscure undesirable
consequences of TFS over the longer term. If healing does
not take place, then glenohumeral joint degeneration may
rival what is seen in delayed surgery of the knee,52 possibly
rendering the joint inoperable,13,17,30,53 as Zingg et al found.30
This is a potential risk of TFS but would be at least several
years in the making.
CONCLUSION
The TFS appears to reduce the pain and disability of RCS
quickly and permanently for some patients. This study
suggests future prospective randomized, controlled, and
double-blinded investigations that may verify a nonsurgi-
cal, low-cost, painless, and virtually immediate means of
treating some cases of RCS.
References
1. Ejnisman B, Andreoli CV, Soares BG, et al. Interventions for
tears of the rotator cuff in adults. Cochrane Database Syst Rev.
2004;(1):CD002758.
2. Green S, Buchbinder R, Hetrick S. Physiotherapy interven-
tions for shoulder pain. Cochrane Database Syst Rev. 2003;(2):
CD004258.
3. McCallister WV, Parsons IM, Titelman RM, Matsen FA III. Open
rotator cuff repair without acromioplasty. J Bone Joint Surg Am.
2005;87(6):1278-1283.
4. Pearsall AW IV, Ibrahim KA, Madanagopal SG. The results of
arthroscopic versus mini-open repair for rotator cuff tears at
mid-term follow-up. J Orthop Surg. 2007;2:24.
5. Moore DR, Cain EL Jr, Schwartz ML, Clancy WG Jr. Allograft re-
construction for massive, irreparable, rotator cuff tears. Paper
presented at: Program and abstracts of the American Orthopae-
dic Society of Sports Medicine Annual Meeting; July 14-17, 2005;
Keystone, Colorado. Reported by Raffy Mirzayan, MD. In: High-
lights of the American Orthopaedic Society for Sports Medicine
(AOSSM) Meeting 2005, Medscape. Accessed September 23, 2007.
6. Park JY, Chung KT, Yoo MJA. Serial comparison of arthroscop-
ic repairs for partial- and full-thickness rotator cuff tears J Ar-
throscopic Relat Surg. 2004;20(7):705-711.
7. Cole B, McCarty LP III, Kang RW, Alford W, Lewis PB, Hayden
JK. Arthroscopic rotator cuff repair: prospective functional out-
come and repair integrity at minimum 2-year follow-up. J Shoul-
der Elbow Surg. 2007;16(5):579-585.
8. Lahteenmaki HE, Hiltunen A, Virolainen P, Nelimarkka O. Re-
pair of full-thickness rotator cuff tears is recommended regard-
less of tear size and age: a retrospective study of 218 patients. J
Shoulder Elbow Surg. 2007;16(5):586-590.
9. Boissonnault WG, Badke MB, Wooden MJ, Ekedahl S, Fye K.
Patient outcome following rehabilitation for rotator cuff repair
surgery: the impact of selected medical comorbidities. J Orthop
Sports Phys Ther. 2007;37(6):312-319.
10. Anderson K, Boothby M, Aschenbrener D, Van Holsbeeck M.
Outcome and structural integrity of arthroscopic rotator cuff re-
pair using 2 rows of fixation: minimum 2-year follow-up. Am J
Sports Med. 2006;34(12):1899-1905.
11. Baysal D, Balyak R, Otto D, Luciak-Corea C, Beaupre L. Func-
tional outcome and health-related quality of life after surgical
repair of full-thickness rotator cuff tear using a mini-open tech-
nique. Am J Sports Med. 2005;33(9):1346-1355.
12. Gerber C, Fuchs B, Hodler J. The results of repair of massive tears
of the rotator cuff. J Bone Joint Surg Am. 2000;82(4):505-515.
13. Gartsman GM, Khan M, Hammerman SM. Arthroscopic repair
of full-thickness tears of the rotator cuff. J Bone Joint Surg Am.
1998;80(6):832-840.
April–June 2011
30/04/11 10:58 AM
 14. Jae CY, Jin HA, Yong SL, Kyoung HK. Magnetic resonance
arthrographic findings of presumed stage-2 adhesive capsu-
litis focus on combined rotator cuff pathology. Orthopedics.
2009;32:2228-2233.
15. Favard L, Bacle G, Berhouet J. Rotator cuff repair. Joint Bone Spine.
2007;74(6):551-557.
16. Jost B, Pfirrmann CW, Gerber C. Clinical outcome after structural
failure of rotator cuff repairs. J Bone Joint Surg Am. 2000;82:304-314.
17. DeFranco MJ, Bershadsky B, Ciccone Y, Yum JK, Iannotti JP.
Functional outcome of arthroscopic rotator cuff repairs: a cor-
relation of anatomic and clinical results. J Shoulder Elbow Surg.
2007;16(6):759-765.
18. Harryman DT, Mack LA, Wang KY, Jackins SE, Richardson ML, Mat-
sen FA III. Repairs of the rotator cuff: correlation of functional results
with integrity of the cuff. J Bone Joint Surg Am. 1991;73:982-989.
19. Klepps S, Bishop J, Lin J, et al. Prospective evaluation of the ef-
fect of rotator cuff integrity on the outcome of open rotator cuff
repairs. Am J Sports Med. 2004;32:1716-1722.
20. Galatz LM, Ball CM, Teefey SA, Middleton WD, Yamaguchi K.
The outcome and repair integrity of completely arthroscopically
repaired large and massive rotator cuff tears. J Bone Joint Surg
Am. 2004;86:219-224.
21. Nové-Josserand L, Walch G, Adeleine P, Courpron P. Effect of
age on the natural history of the shoulder: a clinical and radio-
logical study in the elderly [in French]. Rev Chir Orthop Repara-
trice Appar Mot. 2005;91(6):508-514.
22. Boileau P, Brassart N, Watkinson DJ, Carles M, Hatzidakis AM,
Krishnan SG. Arthroscopic repair of full-thickness tears of the
supraspinatus: does the tendon really heal? J Bone Joint Surg.
2005;87(6):1229-1240.
23. Oh JH, Kim SH, Ji HM, Jo KH, Bin SW, Gong HS. Prognostic fac-
tors affecting anatomic outcome of rotator cuff repair and corre-
lation with functional outcome. Arthroscopy. 2009;25(1):30-39.
24. Iyengar BKS. Light on Yoga. Rev ed. New York, NY: Schocken
Books; 1976:187.
25. Fishman LM, Saltonstall E, Genis, S. Understanding and pre-
venting yoga injuries. Int J Yoga Ther. 2009;18(1):1-8.
26. Walton J, Mahajan S, Paxinos A, et al. Diagnostic values of tests
for acromioclavicular joint pain. J Bone Joint Surg Am. 2004;86-
A(4):807-812.
27. Park HB, Yokota A, Gill HS, El Rassi G, McFarland EG. Di-
agnostic accuracy of clinical tests for the different degrees of
subacromial impingement syndrome. J Bone Joint Surg Am.
2005;87(7):1446-1455.
28. McCarthy M Jr, Chang CH, Pickard AS, et al. Visual analog scales
for assessing surgical pain. J Am Coll Surg. 2005;201(2):245-252.
29. Goldberg BA, Nowinski RJ, Matsen FA III. Outcome of non-
operative management of full thickness rotator cuff tears. Clin
Orthop Relat Res. 2001;(382):99-107.
30. Zingg PO, Jost B, Sukthankar A, Buhler B, Pfirrmann CW, Ger-
ber C. Clinical and structural outcomes of nonoperative manage-
ment of massive rotator cuff tears. J Bone Joint Surg Am. 2007;
89(4):1928-1934.
31. Ainsworth R. Physiotherapy rehabilitation in patients with
massive, irreparable rotator cuff tears. Musculoskeletal Care. 2006;
4(3):140-151.
32. Ainsworth R, Lewis JS. Exercise therapy for the conservative
management of full thickness tears of the rotator cuff: a system-
atic review. Br J Sports Med. 2007;41(4):200-210.
33. Cummins CA, Murrell GA. Mode of failure for rotator cuff repair
with suture anchors identified at revision surgery. J Shoulder
Elbow Surg. 2003;12(2):128-133.
Topics in Geriatric Rehabilitation
Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
TGR2702-09.indd 161
34. Musil D, Sadovsky P, Stehlik J. Massive tears of rotator cuff—
comparison of mini-open and arthroscopic techniques. Part 1:
Mini—open technique [in Czech.]. Acta Chir Orthop Traumatol
Cech. 2006;73(6):387-393
35. Ozbaydar MU, Tonbul M, Yurdoglu C, Yalaman O. Arthroscop-
ic-assisted mini-open repair of rotator cuff tears [in Turkish].
Acta Orthop Traumatol Turc. 2005;39(2):121-127.
36. Maynou C, Mehdi N, Cassagnaud X, Audebert S, Mestdagh H.
Clinical results of arthroscopic tenotomy of the long head of the
biceps brachii in full thickness tears of the rotator cuff without
repair: 40 cases [in French]. Rev Chir Orthop Reparatrice Appar
Mot. 2005;91(4):300-306.
37. Boehm TD, Werner A, Radtke S, Mueller T, Kirschner S, Gohlke
F. The effect of suture materials and techniques on the outcome
of repair of the rotator cuff: a prospective, randomised study. J
Bone Jt Surg Br. 2005;87(6):819-823.
38. Knudsen HB, Gelineck J, Sojbjerg JO, Olsen BS, Johannsen HV,
Sneppen O. Functional and magnetic resonance imaging evalu-
ation after single-tendon rotator cuff reconstruction. J Shoulder
Elbow Surg. 1999;8:242-246.
39. Gazielly DF, Gleyze P, Montagnon C. Functional and anatomical re-
sults after rotator cuff repair. Clin Orthop Relat Res. 1994;304:43-53.
40. Van Linthoudt D, Deforge J, Malterre L, Huber H. Rotator cuff
repair. Long-term results. Joint Bone Spine. 2003;70(4):271-275.
41. Baydar M, Akalin E, El O, et al. The efficacy of conservative
treatment in patients with full-thickness rotator cuff tears [pub-
lished online ahead of print October 12, 2008]. Rheumatol Int.
2009;29(6):623-628. doi: 10.1007/s00296-008-0733-2.
42. Burkhart SS, Barth JR, Richards DP, Zlatkin MB, Larsen M. Ar-
throscopic repair of massive rotator cuff tears with stage 3 and
4 fatty degeneration. Arthroscopy. 2007;23(4):347-354.
43. Ebell MH. Diagnosing rotator cuff tears. Am Fam Phys. 2005;71(8):
1587-1588.
44. Kamath G, Galatz LM, Keener JD, Teefey S, Middleton W,
Yamaguchi K. Tendon integrity and functional outcome after
arthroscopic repair of high-grade partial-thickness supraspina-
tus tears. J Bone Joint Surg Am. 2009;91(5):1055-1062.
45. Hansen ML, Otis JC, Johnson JS, Cordasco FA, Craig EV, Warren
RF. Biomechanics of massive rotator cuff tears: implications for
treatment. J Bone Joint Surg Am. 2008;90(2):316-325.
46. Karduna AR, Kerner PJ, Lazarus MD. Contact forces in the sub-
acromial space: effects of scapular orientation. J Shoulder Elbow
Surg. 2005;14(4):393-399.
47. Di Mario M, Fraracci L. MRI study of the intrinsic acromial angle
in 74 symptomatic patients. Radiol Med. 2005;110(3):273-279.
48. Sharkey NA, Marder RA, Hanson PB. The entire rotator cuff con-
tributes to elevation of the arm. J Orthop Res. 1994;12(5):699-708.
49. Hughes RE, An KN. Force analysis of rotator cuff muscles. Clin
Orthop Relat Res. 1996;(330):75-83.
50. Scibek JS, Mell AG, Downie BK, Carpenter JF, Huges RE. Shoulder
kinematics in patients with full-thickness rotator cuff tears after a
subacromial injection. J Shoulder Elbow Surg. 2008;17(1):172-181.
51. Steenbrink F, de Groot JH, Veeger HE, Meskers CG, van de
Sande MA, Rozing PM. Pathological muscle activation patterns
in patients with massive rotator cuff tears, with and without
subacromial anaesthetics. Man Ther. 2006;11(3):231-237.
52. Jones HP, Appleyard RC, Mahajan S, Murrell GA. Meniscal and
chondral loss in the anterior cruciate ligament injured knee.
Sports Med. 2003;33(14):1075-1089.
53. Laudicina L, D’Ambrosia R. Management of irreparable rotator
cuff tears and glenohumeral arthritis. Orthopedics. 2005;28(4):
382-388.
www.topicsingeriatricrehabilitation.com 161
30/04/11 10:58 AM